EP4053650B1 - Procédé mis en uvre par ordinateur, dispositif de traitement des données et système informatique permettant de commander un dispositif de régulation d'un système de transport - Google Patents

Procédé mis en uvre par ordinateur, dispositif de traitement des données et système informatique permettant de commander un dispositif de régulation d'un système de transport Download PDF

Info

Publication number
EP4053650B1
EP4053650B1 EP21159819.8A EP21159819A EP4053650B1 EP 4053650 B1 EP4053650 B1 EP 4053650B1 EP 21159819 A EP21159819 A EP 21159819A EP 4053650 B1 EP4053650 B1 EP 4053650B1
Authority
EP
European Patent Office
Prior art keywords
piece goods
action
elements
state
vectors
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
EP21159819.8A
Other languages
German (de)
English (en)
Other versions
EP4053650A1 (fr
Inventor
Dr. Clemens OTTE
Dr. Daniel HEIN
Dr. Marc Christian WEBER
Michael Zettler
Dr. Frank Pfeiffer
Martin SCHALL
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Koerber Supply Chain Logistics GmbH
Original Assignee
Koerber Supply Chain Logistics GmbH
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to EP21159819.8A priority Critical patent/EP4053650B1/fr
Application filed by Koerber Supply Chain Logistics GmbH filed Critical Koerber Supply Chain Logistics GmbH
Priority to DK21159819.8T priority patent/DK4053650T3/da
Priority to PT211598198T priority patent/PT4053650T/pt
Priority to ES21159819T priority patent/ES2966145T3/es
Priority to EP22706264.3A priority patent/EP4302159B1/fr
Priority to PCT/EP2022/052335 priority patent/WO2022184358A1/fr
Priority to US18/548,293 priority patent/US20240140724A1/en
Priority to CA3212285A priority patent/CA3212285A1/fr
Publication of EP4053650A1 publication Critical patent/EP4053650A1/fr
Application granted granted Critical
Publication of EP4053650B1 publication Critical patent/EP4053650B1/fr
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
    • B65G43/00—Control devices, e.g. for safety, warning or fault-correcting
    • B65G43/08—Control devices operated by article or material being fed, conveyed or discharged
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
    • B65G43/00—Control devices, e.g. for safety, warning or fault-correcting
    • B65G43/10—Sequence control of conveyors operating in combination
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65G—TRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
    • B65G47/00—Article or material-handling devices associated with conveyors; Methods employing such devices
    • B65G47/22—Devices influencing the relative position or the attitude of articles during transit by conveyors
    • B65G47/26—Devices influencing the relative position or the attitude of articles during transit by conveyors arranging the articles, e.g. varying spacing between individual articles
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/4189—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by the transport system

Definitions

  • the present invention relates to the technical field of conveyor systems for piece goods, in particular conveyor systems that are suitable for singulation and/or alignment of the piece goods.
  • singulators are used to separate an incoming stream of large numbers of unsorted piece goods, in particular postal items such as parcels and parcels or pieces of luggage, i.e. to create a defined distance between individual packages, and often also to ensure a specific alignment of the piece goods. This is necessary in order to be able to process the piece goods in downstream process steps, e.g. for scanning addresses.
  • Another goal is to maximize the throughput (packets/hour) while maintaining a certain control quality (distances and alignment) and possibly other secondary conditions such as reducing power consumption and wear.
  • singulators with conveyor sections running parallel, each of which has a plurality of conveyor elements arranged one behind the other, in which the position and alignment of the piece goods is monitored with sensors (eg cameras).
  • a singulator is controlled by a control device.
  • Piece goods which are brought onto the singulator as a random piece goods stream, are to be transported by the conveyor elements and, at the same time, separated and aligned at defined distances.
  • the conveyor belts of all conveying elements can be controlled separately, with the target values for the speeds being specified by a controller.
  • the control processes are carried out under test conditions, before installation at the end customer, set with a standard flow of goods with a specific fixed distribution of cargo properties.
  • the individual control of the conveying speeds of all conveying elements predetermined for this arrangement is selected by the controller and the conveying elements are controlled accordingly, ie the conveying elements are accelerated and decelerated differently.
  • the optimal, manual or manually-supported setting of these control processes is very complex, because in order to achieve efficient separation and arrangement, the speeds of the conveying elements have to be reset with a very high cycle (eg 30ms).
  • this presetting of the control processes is only very efficient and reliable for the separation and alignment if the piece goods actually transported on the conveyor system have similar properties (weight distribution, friction properties, size, shape, material, ...) as the standard goods used for the presetting exhibit. If, however, the spectrum of goods has different properties than standard goods (e.g. more smooth, slippery plastic packets instead of handy cardboard packets or packets), piece goods do not react to a change in parameters like standard goods. These properties are not necessarily observable directly in the camera image, but they do influence the dynamics, for example through a changed dead time when the speed changes. Adaptations to customer-specific situations after commissioning are difficult, especially if the properties of the piece goods flow at the respective customer change over time after the system has been commissioned.
  • standard goods e.g. more smooth, slippery plastic packets instead of handy cardboard packets or packets
  • EP 3 287 396 A1 discloses a conveyor system for transporting piece goods with control of the distance between piece goods. An iterative adaptive algorithm is used to detect the position of the objects.
  • the object of the present invention is therefore to provide a method and a device which offer an improvement over the prior art. This object is solved by the solutions described in the independent claims. Advantageous configurations result from the dependent claims.
  • the solution according to the invention relates to a device for data processing for a computer-implemented control of a control device of a conveyor system for transporting piece goods of at least one type, in particular postal items and pieces of luggage.
  • the conveying system has a plurality of conveying elements aligned along and parallel to a conveying direction, the conveying elements being driven under the control of the control device by a respectively assigned drive at an individually adjustable speed in order to achieve an alignment and/or a defined spacing of the piece goods.
  • the activation of the control device is determined by at least one agent acting according to methods of reinforcement learning, which according to a strategy for piece goods of a type situationally selects an action from an action space for an initial state in order to reach a subsequent state, the states with state vectors and the Actions with action vectors can be mapped, with the piece goods on the conveyor system can be detected by at least one sensor and the control device comprises a computing unit.
  • the device includes means for carrying out the method according to the invention.
  • the solution according to the invention also relates to a conveyor system for transporting piece goods of at least one type, in particular mail items and items of luggage, which has a plurality of conveyor elements aligned along and parallel to a conveying direction.
  • the conveying elements are driven under the control of a control device by a respectively assigned drive at an individually adjustable speed in order to achieve an alignment and/or a defined spacing of the piece goods.
  • the activation of the control device is determined by at least one agent acting according to methods of reinforcement learning, which, according to a strategy that is the same for all piece goods of a type, chooses an action from an action space for an initial state in order to reach a subsequent state, the states with state vectors and the actions can be mapped with action vectors, comprising a device according to the invention.
  • the solution according to the invention also relates to a computer program comprising instructions which, when executed by a processing unit connected to a conveyor system according to the invention, cause the latter to carry out the method according to the invention.
  • the device, the conveyor system and the computer program have the same advantages, insofar as they can be transferred, which are listed with regard to the method presented.
  • the dimension of each state vector is predetermined by the prior definition of the dimension of the state vectors of all piece goods and the dimension of the state vectors of all piece goods agrees.
  • the dimension of the action vectors is smaller than the number of conveyor elements.
  • the action vector represents the available motor function of the conveying elements and there is no change if a piece of goods is already perfectly aligned and at the desired distance from the neighboring piece of goods.
  • the action vectors can be preassigned with default values of the conveyor system.
  • the agent each piece goods type
  • the number of piece goods and the number of types of piece goods do not change the principle of the process.
  • the agent learns faster because the strategy is not only trained once with just one pass, but is trained according to the number of piece goods and thus optimized faster. As a result, the method can adjust particularly quickly to a changing piece goods flow.
  • a conveyor element By adjusting the speed, a conveyor element is accelerated or decelerated via its drive, as a result of which the conveyor elements change the orientation and position of the piece goods resting on them.
  • standard values can be assigned to all conveying elements of the piece goods.
  • the images are obtained via a camera image (image sensor) and/or via other sensors for determining the orientation and position of the piece goods and are converted into an image that can be described by state vectors.
  • the piece goods in the image can be assigned to a first and at least one other type depending on the properties of the piece goods and, for each assigned type, an agent can be provided with a strategy for piece goods of this type. If for all piece goods in the image If the same strategy is used, all general cargo belongs to one type and there is no assignment of general cargo. However, if easily distinguishable types of general cargo (e.g. cardboard packages as general cargo of the first type and plastic bag packages as general cargo of the second type; rigid suitcases as general cargo of the first type and flexible travel bags as general cargo of the second type; ...) are transported on the same conveyor system, then These piece goods have different adhesion and friction properties.
  • general cargo e.g. cardboard packages as general cargo of the first type and plastic bag packages as general cargo of the second type; rigid suitcases as general cargo of the first type and flexible travel bags as general cargo of the second type;
  • the conveyor system can determine the type of piece goods, eg based on the image, and then allocate a separate strategy to each assigned piece goods type, ie strategy one for cardboard packages and strategy two for plastic bags and any other strategies for other types of piece goods.
  • the speeds of those conveying elements on which the piece good rests but on which no action vector of this piece good has been mapped can be determined and these same conveying elements can be correspondingly individually controlled with the control device.
  • the speeds can be determined by interpolating the speeds of those adjacent conveyor elements onto which an action vector of this item has been mapped. This solves the problem that the dimensionality of the action vector does not necessarily correspond to the number of conveyor elements on which the piece goods rest. Bilinear interpolation, for example, is suitable for this interpolation task.
  • the speeds of all those conveyor elements on which no piece goods are lying and on which no action vector of a piece goods has been mapped can be determined, and corresponding individual ones Control of these conveyor elements with the control device.
  • the speeds can be determined via interpolation, for example bilinear interpolation, of the speeds of those adjacent conveyor elements onto which an action vector of a piece good has been mapped. Special boundary conditions can be assumed for edge conveyor elements. Additionally or alternatively, the speeds can be determined using speed parameters of the conveyor system. These can be standard values from the installation or simulation, eg the mean value of all action vector conveying elements.
  • the speed of the conveying elements on whose adjacent conveying elements the action vector of a piece good has been mapped can be selected such that they match the speed of this adjacent conveying element.
  • the speeds for some or all of these conveying elements can be identical and can be determined from the mean value of the speeds of the conveying elements on which an action vector of a piece good has been mapped.
  • the status information of a piece good that is depicted in the status vector can include position and/or alignment.
  • the status information mapped in the status vector or otherwise mapped status information of a piece good can also include Overlap of the piece goods with those conveyor elements on which the piece goods rest and/or status information of a predetermined number of nearest neighboring pieces of goods within a predetermined distance, at least including their position and/or distance to the piece goods of the state vector, with a smaller number than the predetermined number of next adjacent piece goods the state vector is assigned standard values; and/or speed and/or size of the cargo; and/or global status information of the conveyor system, for example comprising a number of piece goods on the conveyor system, average speed of the conveyor system, prioritization of individual piece goods, for example based on size and/or sorting criterion.
  • the standard values can, for example, depict virtual, already perfectly aligned piece goods at the desired distance, so that these virtual piece goods have as little disruptive influence as possible on the control of the piece goods under consideration.
  • the actual number of belts that a unit load rests on varies depending on the size and orientation of the unit load.
  • the action vectors have a constant dimension.
  • the action vector can only describe velocities that are below predetermined points or surface areas of the piece goods. For example, the corner points of a circumscribing rectangle and/or an approximate center of gravity are well suited as predetermined points.
  • the location and position of the piece goods is thus abstracted and determined by a selection of support parameters that are selected in such a way that they can be influenced with the action vectors.
  • the actual piece goods are always described with a fixed number of parameters with regard to their position on the conveyor system.
  • the properties of the piece goods have been abstracted in with the action vector influenceable model parameters whose number corresponds to the dimensionality of the action vector.
  • this conveying element can be controlled with an average value of these elements or one of the elements can be given full or weighted preference.
  • the image can be evaluated using image processing methods and the state vectors can be created based on the evaluated image.
  • the piece goods are simulated, for example, with circumscribing rectangles.
  • a first attempt to create the state vectors can be made automatically from the image using deep reinforcement learning.
  • a first or further attempt to create the state vectors based on the original image can therefore be made.
  • the representation of the state vectors is not predefined, but learned automatically from the (camera) image using deep reinforcement learning; the state vectors thus become direct based on the pixel occupancy of the digital camera image.
  • the state vectors are determined via the intermediate step of image processing methods carried out on the image, the state vectors are defined by expert knowledge, and errors in image processing also have a direct effect on the state vectors. If for some reason this first attempt to create an image or part of an image is unsuccessful, an attempt can then be made to evaluate the state vectors for this image or part of this image using image processing methods in order to obtain the state vectors.
  • the learning agent can optimize its strategy based on the comparison of the state vectors of the initial state and the subsequent state using a reward and adjust the action vectors of the action space.
  • the agent's strategy for piece goods of a type can be trained with a virtual conveyor system (and thus virtual piece goods and virtual conveyor elements) or with a real conveyor system. If a predetermined strategy is already supplied, it is also possible to dispense with training the strategy while the method is being carried out, for example in the case of a very similar flow of piece goods or insufficient computing capacity in the IT system of the conveyor system operator.
  • figure 1 shows a corresponding conveyor system 2, which transports piece goods 4 along a main conveying direction 6 on a conveying path 8 resting on conveying means 12, with a typical area of use as a singulator 2 in the postal and logistical area.
  • the conveying means 12 are arranged parallel to the main conveying direction 6 in segments 10 arranged one behind the other along the main conveying direction 6, aligned and along a line.
  • the piece goods 4 are transferred from one segment 10 to the respective following segment 10 and lie on several conveyors 12 at the same time and can therefore be separated and/or rotated during their transport if the conveyors 12 are individually controlled by a control device (not shown here).
  • the control device comprises a computing unit that is not shown in the figure.
  • the conveyor system 2 comprises a plurality of sensors 26 arranged above and along the conveyor path and designed as optical detectors. In principle, however, other types of sensors can also be used, as long as the processing unit is able to use the sensor input to determine the state vectors of the piece goods 4 to create. In principle, a single sensor 26 can already be sufficient if the viewing angle is good.
  • the conveyor system 2 is subdivided into segments 18 , 20 , 22 , 24 that perform essentially different tasks along the main conveying direction 6 .
  • a widening device 18 based on the arrangement of the conveying elements 12 to widen the distribution of piece goods. Only one transport along the main conveying direction 6 then takes place on a transfer conveyor 20 .
  • the segments 10d-10h or their conveying means 12 in the alignment section 22 are relatively short.
  • the segments 10d - 10h or their conveyor means 12 in the distance correction section 24 are longer than those of the alignment section 22. It is possible to divide the sections 22, 24 of the conveyor system 2 into sub-conveyor systems with different strategies (higher reward for good alignment in the alignment section 22 or for well-adjusted distances in the distance correction section 24), so that a strategy that is optimized or can be optimized for the relevant section 22, 24 is used. However, this procedure of dividing into different sections 22/24 is particularly suitable for support systems 2 that do not use reinforcement learning methods. According to one embodiment, a reward is also given based on a comparison of the state vectors of the initial and subsequent states s n (t), s n (t+ ⁇ t) in order to achieve an even better and faster optimization of the strategy.
  • the optimal control behavior of the control device of the conveyor system 2 is learned by machine using reinforcement learning ( figure 4 ).
  • an "agent” interacts with the environment, which can either be a concrete system as a conveyor system 2, its simulation/digital twin or a data-driven learned model ("surrogate model") of the system 2 or simulation.
  • the actions with which the environment is influenced are the speeds v of all conveyor elements 12 (eg conveyor belts) and are represented as available motor functions in action vectors a n (t) with typically lower dimensionality than the number of conveyor elements 12 .
  • Observations that are available to the agent as input data are images of the conveyor system, based in particular on cameras 26 and/or other sensor data, and are represented in state vectors s n (t).
  • the action vector will depict a simple onward transport in the conveying direction 6.
  • the agent's behavior is optimized using a reward signal that describes the quality of the current situation. Essentially, the quality is determined by the position/alignment and the distances between the packages 4. For example, the reward value is high when the packages 4 have a defined target distance from one another and are at a specific angle on the conveyor system 2 or its conveyor elements 12 .
  • power consumption, lifetime consumption, noise emissions, etc. can also be taken into account as rewards.
  • the piece goods 4 on the image are assigned to a first and at least one further type depending on properties of the piece goods 4.
  • An agent will provide a separate strategy for each assigned type of piece goods. If only one strategy is used for all piece goods 4, no assignment has to be done.
  • the piece goods 4 are assigned to a type depending on the properties of the piece goods.
  • the assignment can be done using the image or determined beforehand (e.g. at a sorting station), in which case the individual piece goods must be precisely tracked during the process in order not to lose the assignment to a piece goods type.
  • Possible properties that determine the assignment to a piece goods type can be category (packages, parcels, large letters, ...), packaging material (cardboard or plastic), weight (since it influences adhesion on the conveyor elements), size (determines on how many A piece of goods rests on conveyor elements) ....
  • the conveyor system determines the type of piece goods 4, e.g. using the image or using additional sensors, and then allocates a separate strategy to each assigned type of piece goods, e.g. strategy one for heavy cardboard packages and strategy two for light cardboard packages , strategy three for heavy plastic bags, strategy four for light plastic bags and any other strategy for other types of general cargo.
  • FIG 2 shows possible arrangements of the conveyor elements 12 of the conveyor system 2.
  • All conveying elements 12 are arranged in a net-like manner in the manner of a matrix. This form is the easiest to describe and the mapping of an action vector a n (t) onto the real conveying elements 12 is thus particularly uncomplicated and always causes a comparable effect through the driven conveying elements 12 than with another arrangement.
  • the conveying elements 12 in Figure 2b are offset in segments transversely to the conveying direction 6, so that two adjacent conveying elements 12 each open into a conveying element 12.
  • the conveying elements 12 arranged one behind the other along the conveying direction 6 each form continuous conveying sections which are each arranged offset with respect to their conveying elements 12 .
  • the arrangement of Figures 2b, 2c can, however, offer advantages in particular for smaller piece goods 4 of a package flow, which otherwise only lie on one conveyor element 12 .
  • figure 3 shows a flow chart for determining the action vector a(t) according to the invention. Since a belt speed from a continuous range (e.g. between 0.1 m/s and 3.5 m/s) must be set for each conveyor element 12, with 85 conveyor elements, for example, the action space is a subset of the R ⁇ 85, which is far above that which can be learned using known methods complexity (eg because in general the number of required training examples increases exponentially with the dimensionality of the data spaces).
  • a belt speed from a continuous range e.g. between 0.1 m/s and 3.5 m/s
  • a state vector s(t) is not created for the entire conveyor system 2, but based on an image of the sensor 26, an individual state vector s 1 (t), s n (t) is created for each item 4 1 , 4 n .
  • the state vectors s 1 (t), s n (t) are constructed in such a way that they have the same dimensionality for each item 4 1 , 4 n . This means that in particular the number of adjacent items 4' considered remains constant, for example by being limited to the next two or three items at a predetermined distance. Items 4 that are further away are irrelevant for the alignment and spacing of this item 4 and do not have to be taken into account.
  • a state vector s n (t) of constant size is obtained independently of the actual number of piece goods 4 .
  • the corresponding status information of the status vector s n (t) can be filled with standard values. Values that come from so-called virtual piece goods 4' with sufficient spacing and perfect alignment on the belt are available here, for example.
  • the values of the virtual piece goods 4 ⁇ should be selected in such a way that they have as little influence as possible on the control of the piece goods 4 n under consideration.
  • the dimension of the action vectors a n (t) is therefore smaller than the number of conveyor elements 12 of the entire conveyor system 2 in order to achieve a reduction in the dimensionality of the overall problem.
  • a suitable abstraction must be found for this.
  • the action vector a n (t) per piece good 4 n can be selected in such a way that it contains only specific conveying elements 12, e.g. centroid v c , ( figure 5 ).
  • a 5-dimensional action vector a n (t) would be given by the belt speeds v 21 , v 11 v 13 , v 23 (2.01, 2.04, 2.04, 0.10) [m/s] under the 4 corner points and by the belt speed v s ( 2.04 m/s) below the center of gravity.
  • An alternative representation of the action vector a n (t) would be to divide the base area of the item 4 n or a circumscribing rectangle into a fixed number of zones, with each zone being described by a speed v i .
  • the action vector a n (t) can also describe a speed vector of the item 4 n .
  • the representation of the action vector a n (t) is in any case independent of the actual conveying elements 12, but determines their activation in the further course of the method.
  • Methods of reinforcement learning use a strategy function (policy), which maps a state vector s n (t) onto an action vector a n (t) of the action space, ie the strategy function selects suitable belt speeds depending on the respective, in the state vector s n ( t) depicted situation.
  • the strategy function is usually represented by a machine-learned model (neural network, Gaussian process, random forest, parameterized equations, etc.).
  • the mapping of the selected action vector a n (t) onto the real conveying elements 12 influences the subsequent state s n (t+ ⁇ t) of the piece goods.
  • a reward based on which the agent adapts the action vectors of the action space and thus improves the strategy. It is also possible, for a comparison of the following state s n (t+ ⁇ t) with the initial state s n (t) or with states s n (t- ⁇ t), s n (t-2At),... a award reward.
  • This isolated comparison of the subsequent state with the previous state or with more than just the immediately preceding state and/or the isolated evaluation of the subsequent state s n (t+ ⁇ t) combined with a reward-quantified evaluation enables the strategy model to be adjusted.
  • the strategy model is thus improved so that in the future even more suitable action vectors a n (t) are selected for the initial state s n (t) and mapped to the real conveyor system 2 .
  • the strategy can be trained and predetermined in advance using training data (e.g. historical data of the operation of the system using the "standard control"), with the same or a comparable system 2 and other piece goods allocation or using a simulation of the system 2.
  • this predetermined strategy can be used as a predetermined "initial strategy” and this predetermined strategy is then used further trained and thus optimized during the implementation of the procedure.
  • this predetermined strategy is simply applied to the states of the piece goods 4 n mapped in the state vectors s n (t) during the runtime without further optimization—the strategy is then no longer changed during the runtime.
  • the states of the piece goods 4 from the real world can be mapped into state vectors s n (t) of the virtual world.
  • An action vector a n (t) is selected individually for each piece of goods 4 on the basis of its state vector s n (t) using a strategy in the virtual world.
  • This action vector a n (t) can in turn be mapped back onto the conveying elements 12 of the real conveying system 2, so that these conveying elements 12 are controlled with the illustrated speeds of the action vector a n (t), whereupon the piece goods 4 and the entire conveying system 2 in a subsequent state is transferred.
  • this process is evaluated using a reward, which improves the strategy. This process is carried out for each item 4 in the area of the image until the item 4 has left the area of the image.
  • the speeds v of these conveyor elements 12 are determined via interpolation, eg bilinear interpolation, of the speeds v of those adjacent conveyor elements 12 onto which an action vector a n (t) of this item 4 n has been mapped.
  • the speeds v of these conveying elements 12 can be determined using one of the following approaches, which can also be combined with one another: Via interpolation of the speeds v of those adjacent conveyor elements 12 on which an action vector a n (t) of a piece good 4 n has been mapped. Special boundary conditions can be assumed for edge conveyor elements 12 .
  • the speeds v are determined using speed parameters of the conveyor system 2 (standard values from installation or simulation, eg mean value of all action vector conveyor elements).
  • the speed v of the conveying elements 12 on whose adjacent conveying elements 12 the action vector a n (t) of a piece good 4 n has been mapped are chosen such that they correspond to the speed of this adjacent conveying element 12 .
  • Potential conflicts can arise here and can be resolved by prioritizing and/or weighted averaging, for example.
  • the speeds for some or all of these conveying elements 12 can be identical and can be determined from the mean value of the speeds of the conveying elements 12 onto which an action vector a n (t) of a piece good 4 n has been mapped.
  • a major advantage of the method according to the invention is that the strategy is trained from the point of view of one item 4 n for all future items 4 (and for future states of this same item 4 n ) and is also used as a common, shared strategy for all items 4 .
  • the same strategy model is therefore applied to each item 4, 4 1 , 4 n and calculates an individual, local action vector a 1 (t), a n (t) based on the individual state vector s 1 (t), s n ( t).
  • the action vectors a 1 (t), a n (t) are then mapped onto the real conveyor elements 12 as a global band matrix (comprising all conveyor elements 12). Intermediate conveying elements 12 receive suitably interpolated values (eg via bilinear interpolation). Conflicts can arise when mapping onto the real belt matrix, ie more than one package 4 is addressed to the same conveyor element 12. These conflicts, of which figure 7 several are shown are resolved by prioritizing and/or weighted averaging depending on the overlap of the piece goods 4 with the conveying element 12 and the state of the package. For example, a package 4 with little overlap receives a low weighting of the speed of its action vector a(t) projected onto the conveying element 12 in the averaging. A corresponding logic can be specified via expert knowledge or can also be learned by machine. The overlap of each item 4 with its conveying elements 12 can be mapped in the state vector s n (t) or in some other way.
  • the strategy function can be trained using real or simulated data.
  • the training can be continued at the customer's site, which enables the conveyor system to adapt to changing properties of the parcel flow (size, weight, shape and material of the parcels) are automatically adapted.
  • the state vector s n (t) of a unit load 4 n include one or more of the following information: State information of the relevant package 4 (and neighboring packages 4 ') such as positions, speeds, orientation, .... Global information about the Status of the conveyor system 2: number of packages 4, average speed v, prioritization by the user, ....

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • General Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Manufacturing & Machinery (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Mechanical Engineering (AREA)
  • Control Of Conveyors (AREA)

Claims (15)

  1. Procédé mis en oeuvre par ordinateur pour piloter un dispositif de régulation d'un système de transport (2) destiné à transporter des articles (4) d'au moins un type, en particulier des envois postaux et des bagages, le système de transport (2) comprenant une pluralité d'éléments de transport (12) orientés le long et parallèlement à une direction de transport (6), les éléments de transport (12) étant entraînés, sous la commande du dispositif de régulation, par un entraînement associé respectif à une vitesse (v) réglable individuellement, afin d'obtenir une orientation et/ou une distance définie des articles,
    caractérisé en ce que
    le pilotage du dispositif de régulation (14) est déterminé par au moins un agent agissant ou prédéterminé selon des méthodes d'apprentissage par renforcement, qui, conformément à une stratégie, sélectionne en fonction de la situation, pour un état initial, une action à partir d'un espace d'action, afin de parvenir à un état suivant, les états pouvant être représentés par des vecteurs d'état -s(t), s(t+ Δt) et les actions pouvant être représentées par des vecteurs d'action -a(t), a(t+ Δt), comprenant les étapes consistant à :
    a) créer une image initiale du système de transport (2) ;
    b) pour chacun des articles (4n) sur l'image, créer individuellement un vecteur d'état sn(t) de dimension définie préalablement et identique pour tous les articles (4, 4n) d'un même type, comprenant des informations d'état de l'article (4n) correspondant prélevées sur l'image créée immédiatement auparavant ;
    c) pour chaque article (4n), sélectionner individuellement un vecteur d'action an(t) à partir d'un espace d'action, selon la stratégie identique pour tous les articles (4) d'un même type, pour le vecteur d'état actuel sn(t) de cet article (4n), la dimension du vecteur d'action an(t) étant prédéterminée ;
    d) pour chaque article (4n), représenter le vecteur d'action an(t) sur les éléments de transport réels (12) dudit article (4n), afin de déterminer la vitesse v desdits éléments de transport (12), et piloter en correspondance les éléments de transport (12) au moyen du dispositif de régulation ;
    e) après écoulement d'un temps de cycle Δt, créer une image suivante du système de transport (2) et exécuter l'étape b) du procédé, afin d'obtenir, pour chaque article (4n), un vecteur d'état de l'état suivant sn(t+ Δt) ;
    f) si la stratégie pour les articles (4) d'un même type doit être entraînée davantage pendant la mise en oeuvre du procédé, pour chaque article (4n) dudit type, évaluer le vecteur d'état de l'état suivant sn(t+ Δt) par une méthode d'apprentissage par renforcement à l'aide d'une récompense, après quoi l'agent entraîne et donc optimise sa stratégie pour les articles (4) dudit type en adaptant les vecteurs d'action an (t) de l'espace d'action ;
    g) pour chaque article (4n), exécuter une nouvelle fois les étapes c) à f) du procédé, en utilisant la stratégie améliorée ou prédéterminée tant que l'article (4n) correspondant est représenté sur l'image suivante.
  2. Procédé selon la revendication 1,
    comprenant en outre l'étape consistant à
    affecter les articles (4, 4') sur l'image à un premier type et à au moins un autre type en fonction des propriétés des articles (4, 4') et, pour chaque type affecté, mettre à disposition un agent avec une stratégie pour les articles (4, 4') dudit type.
  3. Procédé selon l'une des revendications 1 ou 2,
    comprenant en outre l'étape consistant à
    pour chaque temps de cycle Δt pour chaque article (4n), déterminer les vitesses v des éléments de transport (12) sur lesquels repose l'article (4n) mais sur lesquels aucun vecteur d'action an(t) dudit article (4n) n'a été représenté, et piloter individuellement en correspondance ces mêmes éléments de transport (12) au moyen du dispositif de régulation, les vitesses v étant déterminées par interpolation des vitesses v des éléments de transport (12) voisins sur lesquels un vecteur d'action an(t) dudit article (4n) a été représenté.
  4. Procédé selon l'une des revendications 1 ou 3,
    comprenant en outre l'étape consistant à
    pour chaque temps de cycle Δt, déterminer les vitesses v de tous les éléments de transport (12) sur lesquels à la fois aucun article (4) ne repose et sur lesquels aucun vecteur d'action an(t) d'un article (4) n'a été représenté, et piloter individuellement en correspondance ces mêmes éléments de transport (12) au moyen du dispositif de régulation,
    dans lequel
    - les vitesses v sont déterminées par interpolation des vitesses v des éléments de transport (12) voisins sur lesquels un vecteur d'action an(t) d'un article (4n) a été représenté ; et/ou
    - les vitesses v sont déterminées à l'aide de paramètres de vitesse du système de transport (2) ; et/ou
    - les vitesses v des éléments de transport (12) sur les éléments de transport (12) voisins desquels le vecteur d'action an(t) d'un article (4n) a été représenté, sont choisies de manière à coïncider avec la vitesse dudit élément de transport (12) voisin ; et/ou
    - les vitesses sont identiques pour certains ou tous ces éléments de transport (12) et sont déterminées à partir de la moyenne des vitesses des éléments de transport (12) sur lesquels un vecteur d'action an(t) d'un article (4n) a été représenté.
  5. Procédé selon l'une des revendications 1 à 4,
    caractérisé en ce que
    les informations d'état d'un article (4n) représentées dans le vecteur d'état sn(t) comprennent la position et/ou l'orientation de l'article (4n).
  6. Procédé selon l'une des revendications 1 à 5,
    caractérisé en ce que
    les informations d'état d'un article (4) représentées dans le vecteur d'état sn(t) ou représentées autrement comprennent
    - le chevauchement de l'article (4) avec les éléments de transport (12) sur lesquels l'article (4) repose ; et/ou
    - des informations d'état d'un nombre prédéterminé d'articles voisins (4) les plus proches à l'intérieur d'une distance prédéterminée, comprenant au moins leur position et/ou leur distance par rapport à l'article (4) du vecteur d'état sn(t), sachant que, dans le cas d'un nombre inférieur au nombre prédéterminé d'articles voisins (4') les plus proches, le vecteur d'état sn(t) est affecté de valeurs standard ; et/ou
    - la vitesse et/ou la taille de l'article (4) ; et/ou
    - des informations d'état globales du système de transport (2), comprenant par exemple un nombre d'articles (4) sur le système de transport (2), une vitesse moyenne du système de transport (2), une priorisation d'articles individuels (4), par exemple en fonction de leur taille et/ou d'un critère de tri.
  7. Procédé selon l'une des revendications 1 à 6,
    caractérisé en ce que
    le vecteur d'action an(t) décrit uniquement des vitesses qui se situent en dessous de points ou de zones de surface prédéterminés de l'article (4).
  8. Procédé selon l'une des revendications 1 à 7,
    caractérisé en ce que
    lorsque les vecteurs d'action an(t), an'(t) affectés à deux ou à plusieurs articles (4, 4') sont représentés sur le même élément de transport (12), une priorisation et/ou une moyenne pondérée des vitesses prescrites par les vecteurs d'action an(t), an'(t) est effectuée en fonction du chevauchement respectif desdits articles (4) avec ledit élément de transport (12) et/ou d'une qualité des vecteurs d'état sn(t)) ; et/ou
    lorsque deux éléments du vecteur d'action an(t) d'un article (4n) sont représentés sur le même élément de transport (12), ledit élément de transport (12) est piloté avec une valeur moyenne desdits éléments, ou l'un des éléments est priorisé en totalité ou de manière pondérée.
  9. Procédé selon l'une des revendications 1 à 8,
    caractérisé en ce que
    l'image est évaluée par des méthodes de traitement d'image, et les vecteurs d'état sn(t) sont créés sur la base de l'image évaluée.
  10. Procédé selon l'une des revendications 1 à 9,
    caractérisé en ce que
    une première tentative de création des vecteurs d'état sn(t) est effectuée automatiquement à partir de l'image au moyen de l'apprentissage par renforcement en profondeur (deep reinforcement learning).
  11. Procédé selon l'une des revendications 1 à 10,
    comprenant en outre l'étape consistant à
    entraîner la stratégie de l'agent pour des articles (4) d'un même type avec un système de transport virtuel ou réel (2', 2).
  12. Dispositif de traitement de données pour un pilotage mis en oeuvre par ordinateur d'un dispositif de régulation d'un système de transport (2) destiné à transporter des articles (4) d'au moins un type, en particulier des envois postaux et des bagages, le système de transport (2) comprenant une pluralité d'éléments de transport (12) orientés le long et parallèlement à une direction de transport (6), les éléments de transport (12) étant entraînés, sous la commande du dispositif de régulation, par un entraînement associé respectif à une vitesse réglable individuellement, afin d'obtenir une orientation et/ou une distance définie des articles, le pilotage du dispositif de régulation étant déterminé par au moins un agent agissant selon des méthodes d'apprentissage par renforcement, qui, conformément à une stratégie pour des articles (4) d'un même type, sélectionne en fonction de la situation, pour un état initial, une action à partir d'un espace d'action, afin de parvenir à un état suivant, les états pouvant être représentés par des vecteurs d'état et les actions pouvant être représentées par des vecteurs d'action, les articles sur le système de transport (2) pouvant être détectés par au moins un capteur (26), et le dispositif de régulation comprenant une unité de calcul ;
    caractérisé par
    des moyens de mise en oeuvre du procédé selon la revendication 1.
  13. Dispositif selon la revendication 12,
    caractérisé en ce que
    le dispositif est conçu pour mettre en oeuvre le procédé selon l'une des revendications 2 à 11.
  14. Système de transport (2) destiné à transporter des articles (4) d'au moins un type, en particulier des envois postaux et des bagages, le système de transport (2) comprenant une pluralité d'éléments de transport (12) orientés le long et parallèlement à une direction de transport (6), les éléments de transport (12) étant entraînés, sous la commande d'un dispositif de régulation, par un entraînement associé respectif à une vitesse réglable individuellement, afin d'obtenir une orientation et/ou une distance définie des articles, le pilotage du dispositif de régulation étant déterminé par au moins un agent agissant selon des méthodes d'apprentissage par renforcement, qui, conformément à une stratégie identique pour tous les articles (4) d'un même type, sélectionne en fonction de la situation, pour un état initial, une action à partir d'un espace d'action, afin de parvenir à un état suivant, les états pouvant être représentés par des vecteurs d'état et les actions pouvant être représentées par des vecteurs d'action, le système de transport (2) comprenant un dispositif selon la revendication 12 ou 13.
  15. Programme d'ordinateur comprenant des instructions qui, lorsqu'elles sont exécutées par une unité de calcul reliée à un système de transport (2) selon la revendication 14, amènent cette dernière à mettre en oeuvre le procédé selon l'une des revendications 1 à 11.
EP21159819.8A 2021-03-01 2021-03-01 Procédé mis en uvre par ordinateur, dispositif de traitement des données et système informatique permettant de commander un dispositif de régulation d'un système de transport Active EP4053650B1 (fr)

Priority Applications (8)

Application Number Priority Date Filing Date Title
DK21159819.8T DK4053650T3 (da) 2021-03-01 2021-03-01 Computerimplementeret metode, anordning til databehandling og computersystem til betjening af en styreanordning i et transportsystem
PT211598198T PT4053650T (pt) 2021-03-01 2021-03-01 Processo implementado por computador, dispositivo para o processamento de dados e sistema de computador para o controlo de um dispositivo de regulação de um sistema de transporte
ES21159819T ES2966145T3 (es) 2021-03-01 2021-03-01 Procedimiento implementado por ordenador, dispositivo para el procesamiento de datos y sistema informático para controlar un equipo de regulación de un sistema de transporte
EP21159819.8A EP4053650B1 (fr) 2021-03-01 2021-03-01 Procédé mis en uvre par ordinateur, dispositif de traitement des données et système informatique permettant de commander un dispositif de régulation d'un système de transport
EP22706264.3A EP4302159B1 (fr) 2021-03-01 2022-02-01 Procédé mis en uvre par ordinateur, dispositif de traitement des données et système informatique permettant de commander un dispositif de régulation d'un système de transport
PCT/EP2022/052335 WO2022184358A1 (fr) 2021-03-01 2022-02-01 Procédé mis en œuvre par ordinateur, dispositif de traitement de données, et système informatique pour l'actionnement d'un dispositif de régulation d'un système convoyeur
US18/548,293 US20240140724A1 (en) 2021-03-01 2022-02-01 Computer-implemented method, apparatus for data processing, and computer system for controlling a control device of a conveyor system
CA3212285A CA3212285A1 (fr) 2021-03-01 2022-02-01 Procede mis en ƒuvre par ordinateur, dispositif de traitement de donnees, et systeme informatique pour l'actionnement d'un dispositif de regulation d'un systeme convoyeur

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
EP21159819.8A EP4053650B1 (fr) 2021-03-01 2021-03-01 Procédé mis en uvre par ordinateur, dispositif de traitement des données et système informatique permettant de commander un dispositif de régulation d'un système de transport

Publications (2)

Publication Number Publication Date
EP4053650A1 EP4053650A1 (fr) 2022-09-07
EP4053650B1 true EP4053650B1 (fr) 2023-09-06

Family

ID=74844754

Family Applications (2)

Application Number Title Priority Date Filing Date
EP21159819.8A Active EP4053650B1 (fr) 2021-03-01 2021-03-01 Procédé mis en uvre par ordinateur, dispositif de traitement des données et système informatique permettant de commander un dispositif de régulation d'un système de transport
EP22706264.3A Active EP4302159B1 (fr) 2021-03-01 2022-02-01 Procédé mis en uvre par ordinateur, dispositif de traitement des données et système informatique permettant de commander un dispositif de régulation d'un système de transport

Family Applications After (1)

Application Number Title Priority Date Filing Date
EP22706264.3A Active EP4302159B1 (fr) 2021-03-01 2022-02-01 Procédé mis en uvre par ordinateur, dispositif de traitement des données et système informatique permettant de commander un dispositif de régulation d'un système de transport

Country Status (7)

Country Link
US (1) US20240140724A1 (fr)
EP (2) EP4053650B1 (fr)
CA (1) CA3212285A1 (fr)
DK (1) DK4053650T3 (fr)
ES (1) ES2966145T3 (fr)
PT (1) PT4053650T (fr)
WO (1) WO2022184358A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4342826A1 (fr) * 2022-09-20 2024-03-27 Schneider Electric Industries SAS Procédé de création d'une commande pour un système de transport

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160239000A1 (en) * 2015-02-12 2016-08-18 Nec Laboratories America, Inc. TS-DIST: Learning Adaptive Distance Metric in Time Series Sets
DE102015117241B4 (de) * 2015-10-09 2018-11-15 Deutsche Post Ag Ansteuerung einer Förderanlage
EP3287396B1 (fr) * 2016-08-25 2023-12-06 Körber Supply Chain Logistics GmbH Systeme d'alimentation a segments
DE102019108126A1 (de) * 2019-03-28 2020-10-01 Weber Maschinenbau Gmbh Breidenbach Vorrichtung zum Fördern von Produkten

Also Published As

Publication number Publication date
WO2022184358A1 (fr) 2022-09-09
CA3212285A1 (fr) 2022-09-09
US20240140724A1 (en) 2024-05-02
EP4053650A1 (fr) 2022-09-07
PT4053650T (pt) 2023-12-12
DK4053650T3 (da) 2023-12-11
EP4302159B1 (fr) 2024-10-02
EP4302159A1 (fr) 2024-01-10
ES2966145T3 (es) 2024-04-18

Similar Documents

Publication Publication Date Title
EP3153242B1 (fr) Commande de transporteur
DE69505319T2 (de) Anordnung zum automatischen Öffnen und Schliessen von Türen
WO2015035525A1 (fr) Procédé d'analyse de la distribution d'objets dans des files d'attente libres
EP3510456B1 (fr) Procédé de montage d'un objet
DE112017004812T5 (de) Automatische synchronisierung von pressmaschinen und robotern
EP4053650B1 (fr) Procédé mis en uvre par ordinateur, dispositif de traitement des données et système informatique permettant de commander un dispositif de régulation d'un système de transport
DE102019107414A1 (de) Verfahren sowie Steuergerät für ein System zum Steuern eines Kraftfahrzeugs
AT506964B1 (de) Verfahren und vorrichtung zur bilderkennung von bewegten objekten, wie etwa in förderanlagen
WO2017076541A1 (fr) Mécanisme de tri à commande distribuée
DE102018220941A1 (de) Auswertung von Messgrößen mit KI-Modulen unter Berücksichtigung von Messunsicherheiten
DE102018116611A1 (de) Verfahren und System zur Steuerung des Materialflusses von Objekten in einem realen Lager
DE102017200580A1 (de) Verfahren zur Optimierung einer Manöverplanung für autonom fahrende Fahrzeuge
WO2017216124A1 (fr) Dispositif de triage et procédé de triage correspondant
EP3825263A1 (fr) Procédé de configuration par ordinateur d'une application d'entraînement régulée d'un système logistique
EP3437745A1 (fr) Trieur matriciel
EP0813986B1 (fr) Méthode de commande de distance d'un véhicule
DE19640149A1 (de) Regler für Niveauregeleinrichtung
DE102021200339A1 (de) Sortiersystem und Verfahren zum Steuern eines Sortiersystems
EP3867176A1 (fr) Procédé et système de chargement d'un véhicule de transport sans conducteur
WO2020001710A1 (fr) Système d'aide à la conduite pour la sélection d'un jeu de paramètres de mouvement
DE102019213896A1 (de) Inkrementelles neuronales Netzwerk mit Encoder-Decoder-Struktur
DE102019107209B4 (de) Sortierspeichereinrichtung und Verfahren zum Betreiben einer Sortierspeichereinrichtung
DE102021208722A1 (de) Konfigurieren eines neuronalen Netzwerks auf äquivariantes oder invariantes Verhalten
WO2024175801A1 (fr) Procédé de commande d'un entrepôt de marchandises suspendues, entrepôt de marchandises suspendues et support de marchandises pour un entrepôt de marchandises suspendues à blocage de voie, à évitement de collision, à transfert d'informations et à adressage multiple
DE102019205080A1 (de) Künstliches neuronales Netzwerk mit verbesserter Ermittlung der Zuverlässigkeit der gelieferten Aussage

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION HAS BEEN PUBLISHED

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: KOERBER SUPPLY CHAIN LOGISTICS GMBH

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20230113

RBV Designated contracting states (corrected)

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: GRANT OF PATENT IS INTENDED

RIC1 Information provided on ipc code assigned before grant

Ipc: B65G 43/10 20060101ALI20230509BHEP

Ipc: B65G 43/08 20060101ALI20230509BHEP

Ipc: B65G 47/26 20060101ALI20230509BHEP

Ipc: G05B 19/418 20060101ALI20230509BHEP

Ipc: G05B 13/02 20060101AFI20230509BHEP

INTG Intention to grant announced

Effective date: 20230602

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE PATENT HAS BEEN GRANTED

P01 Opt-out of the competence of the unified patent court (upc) registered

Effective date: 20230720

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

Free format text: NOT ENGLISH

REG Reference to a national code

Ref country code: CH

Ref legal event code: EP

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

Free format text: LANGUAGE OF EP DOCUMENT: GERMAN

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 502021001407

Country of ref document: DE

REG Reference to a national code

Ref country code: DK

Ref legal event code: T3

Effective date: 20231205

REG Reference to a national code

Ref country code: PT

Ref legal event code: SC4A

Ref document number: 4053650

Country of ref document: PT

Date of ref document: 20231212

Kind code of ref document: T

Free format text: AVAILABILITY OF NATIONAL TRANSLATION

Effective date: 20231204

REG Reference to a national code

Ref country code: NL

Ref legal event code: FP

REG Reference to a national code

Ref country code: LT

Ref legal event code: MG9D

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: GR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20231207

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

Ref country code: RS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

Ref country code: NO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20231206

Ref country code: LV

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

Ref country code: LT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

Ref country code: HR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

Ref country code: GR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20231207

Ref country code: FI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240106

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: NL

Payment date: 20240320

Year of fee payment: 4

REG Reference to a national code

Ref country code: ES

Ref legal event code: FG2A

Ref document number: 2966145

Country of ref document: ES

Kind code of ref document: T3

Effective date: 20240418

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SM

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

Ref country code: RO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

Ref country code: IS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240106

Ref country code: EE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

Ref country code: CZ

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

Ref country code: SK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: DE

Payment date: 20240103

Year of fee payment: 4

Ref country code: PT

Payment date: 20240222

Year of fee payment: 4

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: PL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: FR

Payment date: 20240320

Year of fee payment: 4

Ref country code: DK

Payment date: 20240321

Year of fee payment: 4

REG Reference to a national code

Ref country code: DE

Ref legal event code: R097

Ref document number: 502021001407

Country of ref document: DE

PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: CH

Payment date: 20240401

Year of fee payment: 4

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: ES

Payment date: 20240417

Year of fee payment: 4

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20230906

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: IT

Payment date: 20240401

Year of fee payment: 4

26N No opposition filed

Effective date: 20240607